It is widely accepted that reactive oxygen species (ROS), such as H2O2 promote tumorigenesis. Consequently, there is a crucial need for a greater understanding of the mechanism of H2O2-mediated cancer development. We have shown that mice lacking one or two copies of the peroxidase Prdx1 die prematurely of hemolytic anemia and multiple cancers. Furthermore, cells and mice lacking Prdx1 show a higher incidence of transformation to H-RasV12 and ErbB-2 and H-RasV12-induced breast cancer. Fibroblasts and mammary epithelial cells lacking Prdx1 show higher phosphorylation of Akt on Ser473 and more inactive (oxidized) forms of the tumor suppressor PTEN. PTEN is readily inactivated via oxidation of the low pKa cysteine in its catalytic site. We found that Prdx1 binding to PTEN is essential to protect it from oxidation-induced inactivation. The Prdx1:PTEN complex is disrupted by H2O2, upon which Akt including its downstream signaling is activated resulting in a decrease in apoptosis. Therefore, our working hypothesis is that H2O2 inactivates PTEN lipid phosphatase activity due to a complex disruption of Prdx1:PTEN, thereby promoting Akt kinase-driven oncogenesis. We further hypothesize that Prdx1:PTEN binding is required for normal PTEN function, including its lipid phosphatase activity, nuclear stability and apoptosis induction. Specific Aim1 will determine the mechanisms by which H2O2 regulates the Prdx1:PTEN interaction to alter PTEN lipid phosphatase activity and Prdx1 peroxidase activity. We will define if H2O2-induced oxidation of either Prdx1 or PTEN disrupts or prevents the Prdx1:PTEN interaction and modulates PTEN lipid phosphatase activity and Prdx1 peroxidase activity. Specific Aim2 will examine if the Prdx1:PTEN interaction promotes tumor suppression by regulating PTEN or Akt dependent mechanisms. To complete this we will examine PTEN-induced tumor suppressive functions including apoptosis upon disrupting the Prdx1:PTEN interactions by using peptide interference in vitro and in vivo. Specific Aim3 will evaluate if nuclear Prdx1 promotes tumor suppression ErbB-2-induced breast cancer in mice. We have found that loss of Prdx1 decreases nuclear PTEN protein levels. We will therefore introduce Prdx1 fused to nuclear localization signal sequence into mammary epithelial cells isolated from Prdx1-/-MMTV-ErbB-2 mice and transplant cells into clear fad pads of NCR nude mice, to observe if nuclear Prdx1 decreases incidence of breast cancer or tumor burden. Mice lacking Prdx1 offer the first mouse model where loss of a peroxidase results in elevation of endogenous H2O2 thereby causing cancer. This mouse model mimics conditions where H2O2 levels are elevated as found in aging, tobacco smoking, ionizing radiation and environmental carcinogens. By studying breast cancer prone mice lacking Prdx1 we should help to find more specific preventive treatments in H2O2-induced breast cancer.